Syllabus Review 3 (1), 2012 : 257 - 279
N
E
S
SYLLABUS
REVIEW
Human& Social Science Series
Reflexive Anaphora in VP-Elliptical Sentences
of ESL Learners in Cameroon
Napoleon Kang Epoge
Department of English, Ecole Normale Supérieure, University of Yaoundé I,
Email: [email protected]
Abstract
This paper discusses the interpretation of reflexive anaphora in VPelliptical sentences by L2 learners of English in Cameroon. The
interpretation is considered within the Binding Theory of Chomsky (1981)
and the Relevance Theory of Sperber and Wilson (1995). In order to
investigate this, a grammatical judgment task (GJT) was designed to elicit
specific answers. The task which consisted of 14 experimental sentences,
with reflexive anaphora in VP-ellipsis in each sentence, was administered
to 128 respondents. Each experimental sentence was followed by two
alternative interpretations of the VP-ellipsis. Findings reveal that these
respondents’ interpretation of reflexive anaphora in VP-elliptical structures
is characterized by strict and sloppy reading even in referential context
favouring strict reading interpretation. The strategies of ‘naïve optimism’,
‘cautious optimism’ and ‘sophisticated understanding’ apparently played
significant roles here.
Keywords: Reflexive anaphora, Verb phrase, Ellipsis, Cameroon English,
Sloppy reading, Strict reading
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Introduction
Reinhart and Reuland (1993) stipulate that reflexives, which
are constrained by Binding Theory, are bound to the
antecedent in the local clause in which they are found (e.g.
Childreni think [that parentsj overwork themselvesj]). In this
sentence, the bracketed clause is the local clause in which
the reflexive themselves and the antecedent NP parents are
found. Consequently, themselves is bound by parents, the NP
within the minimal binding domain, and not by children, the
NP outside the local domain. Thus, the Binding Theory
(Chomsky 1981) proposes the conditions under which
different nominal expressions establish reference. These
conditions are known as the binding principles.
Principle A: An anaphor must be bound in its minimal
domain;
Principle B: A pronoun must be free in its minimal domain;
Principle C: A referring expression must be free.
Principle A holds that anaphors (reflexives and
reciprocals) are bound within a specific syntactic domain
(their governing category). That is, they have an
antecedent which c-command them within their
governing category (e.g Paul i thinks Mabel j hates
herself j). Pronouns (Principle B) are not bound within the
governing categories; they can be bound only by
elements outside of the governing category (e.g. Jacob i is
a teacher. He i teaches mathematics). All other NPs
(Principle C) are always free (e.g. Jonathan i is a teacher
and Mary j a medical doctor). Violations of these
conditions can be found in the following sentences:
*“Johni thinks that Paul j is buying himself i a picture”
where there is intended co-reference between John and
himself. *“He i thinks that Paul i is buying John j a picture”,
where there is co-reference between he and Paul.
258
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
As can be seen above, Principle A affects reflexives.
A reflexive is a noun phrase which is not interpreted
semantically in its own right but which instead makes
reference to a determiner phrase (henceforth DP) for its
interpretation. The dependency relation of a reflexive to a
DP is known as binding and the DP with the fixed
meaning is the antecedent that binds the reflexive.
(1) Suzy hurt herself.
In this example herself is a reflexive and Suzy the DP.
Consequently, the reflexive herself is a lexical item which
has no fixed meaning but instead makes reference to Suzy
for its interpretation. The dependency relation of the
reflexive herself to the DP Suzy is known as binding and
the DP Suzy with the fixed meaning is the antecedent that
binds the reflexive. In this respect, the proper name Suzy,
to which the reflexive herself points back to, is the
antecedent of the reflexive. This shows that the
antecedent and the reflexive point to the same entity.
This situation whereby the reference expression
(reflexive) and the referent (antecedent) point to the same
entity is known in the literature as co-reference.
Not every DP is a potential binder of a reflexive. In
this connection, Chomsky (1981) stipulates that a
reflexive must be bound within its minimal domain;
where minimal domain is understood to mean the
smallest clause containing the reflexive and its
antecedent as in the example below.
(2) Peter i told [John j to defend himselfj ].
In this example, the reflexive himself and the antecedent
John are in the same clause (clause-mate). Consequently,
the NP John (and not Peter) antecedes the reflexive
259
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
himself in the above sentence. This is indicated in the
above sentence by giving the same index to the
antecedent John and the reflexive himself.
Reinhart and Reuland (1993) also underscore the
fact that logophoric reflexives (i.e., long-distance bound
reflexives or reflexives that are bound to the subject of
the higher clause) are not constrained by Binding
Theory, but are constrained by pragmatic/discourse
factors, and may behave as co-referential anaphora (i.e.,
like pronouns) as illustrated below.
(3) Alfred thinks he is a great cook, and Felix does [e] too.
(a) Alfred x thinks that X is a great cook, and Felix y
<thinks that Y is a great cook>.
(b) ) Alfred thinks that he is a great cook, and Felix
<thinks that he is a great cook>.
(Goldwater et al 2006:2)
As can be inferred above, (3) is ambiguous between the
interpretation in which the elided phrase contains a
variable bound by the immediate subject of the elided
verb phrase (henceforth VP) (3a) or a pronoun coreferential with the subject of the previous sentence
(3b). In this case, the discourse binding constraint states
that the antecedent for a reflexive must be a prominent
discourse NP. This idea of prominent NP is illustrated
in the tree diagram of the following example.
(4) John talked to Paul about himself.
260
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
In this tree diagram, Paul does not c-command himself
since the first branching node dominating Paul is P, which
does not dominate the reflexive himself. However, the verb
talk, has three theta-roles (θ-roles) to assign: the talker, the
person talked to, and the topic of conversation. Thus,
adjoining the second prepositional phrase to Paul prevents
talk from properly assigning the ‘topic of conversation’ θrole to about himself which it should clearly receive.
Consequently, assigning this θ-role to Paul, enables himself
to be bound by Paul. This is consistent with the approach to
binding that takes the antecedent of a reflexive to be the
theta-role that satisfies the binding function introduced by
the reflexive. Furthermore, the NP John c-commands the
reflexive himself since the first branching node dominating
John is a minimal inflectional phrase which also dominates
the reflexive himself. In this way, himself is also bound by the
subject NP John.
261
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Unlike pronouns, Reinhart and Reuland (1993) hold
that reflexives are bound to the local antecedent only. This
belief is constrained by the Closeness Hierarchy Condition
(Pan 1998) which states that the NP closer to the reflexive is
the more prominent NP to antecede the reflexive.
Consequently, the co-referential reading in which the elided
reflexive is co-referential with the subject of the previous
sentence is wanting as exemplified by (5b).
(5) Lucie praised herself, and Lili did [e], too.
a) Luciex praised x, and Liliy <praised y>.
b) ?? Luciei praised herselfi, and Lili <praised herself*i>.
(Goldwater et al 2006:2)
Furthermore, unlike local binding reflexives, logophoric
reflexives behave as pronouns. This is as a result of the fact
that they can act as locally bound anaphors or co-referential
long-distance anaphors as illustrated by (6b).
(6) Lucie liked the picture of herself, and Lili did [e], too.
a) Luciex liked the picture of x, and Liliy<liked the picture of y>.
b) Luciei liked the picture of herselfi, and Lili <liked the
picture of herselfi>.
As the ongoing discussion reveals, reflexive anaphora in
VP-elliptical sentences is interesting in that it allows two
possibilities of interpretation (Fiengo and May 1994). This
view is explicitly illustrated in (7).
(7) John defended himself and Paul did too.
a) Johni defended himselfi and Paulj did [defend himselfj]
too. (i.e., Paul defended Paul)
b) Johni defended himselfi and Paulj did [defend himi]
too. (i.e., Paul defended John).
In (7a), the reflexive himself in the elided VP co-refers with
Paul. This interpretation, known in the literature as sloppy
reading, is set by the requirement of principle A of the
262
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Binding Theory (i.e., an anaphor must be bound in its local
domain; where local domain is understood to mean the
clause containing the anaphor and its antecedent). In (7b),
the pronoun him in the elided VP co-refers with John, the
subject of the higher clause. This interpretation, known as
the strict reading, is set by principle B of the Binding Theory
(i.e., a pronoun must be free in its local domain).
However, the assumption in linguistic literature that
reflexives function obligatorily as bound variables (i.e. as
reflexives binding to the antecedent NP in the local
domain), makes it relatively uncontroversial for them to
have a sloppy reading. This is evidenced by the fact that
reflexive anaphora in English are bound to the closest
antecedent as shown in (8).
(8) Johni defended himselfi and Billj did [defended himselfj] too.
In this example, himself co-refers with John while did too corefers with Bill. It is in this light that reflexives are said to
function obligatorily as bound variables. That is, they are
bound to the closest antecedent.
Though this assumption has been unanimously
accepted, there has been disagreement, in the literature, on
the status of a strict reading. Dalrymple et al (1991)
proposed that the availability of a strict reading of a
reflexive anaphora in VP-elliptical sentences depends on the
semantic property of individual verbs. In this wise, they
distinguish between verbs such as lock and depend, stating
that no matter what the structure is, the verb defend allows a
strict reading whereas the verb lock does not.
263
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
(9) Bill defended himself against the accusation, and John did, too.
(10) John locked himself in the bathroom when bad news
arrived, but Bill would never do so.
They explained that defend gets both a sloppy and strict
reading because it does not intrinsically impose a
requirement of co-reference between its subject and object.
On the other hand, lock does not allow a strict reading
because it imposes a requirement of co-reference between
its subject and object. Though this argument is plausible, it
is not entertained by Hestvik (1995) who points out that the
semantic properties of lexical items do not determine
whether reflexives allow a strict reading or not. Hestvik
compares sentences like (11) and (12) to drive home the
argument.
(11) John locked himself in the bathroom when bad news
arrived, but Bill would never do so.
(12) John locked himself in the bathroom before Bill could.
As can be inferred above, it is evidently revealed that a strict
reading is possible in (12), suggesting that an account based
on the semantic property of verb is wanting (Hestvik 1995).
Furthermore, Kitagawa (1991) approached the problem of
strict reading of reflexives by reconstructing reflexives as
pronouns at the logical form (henceforth LF). She stipulates
that a feature [+anaphor] on the reflexive can be suppressed
in the copying of the antecedent VP into the elided VP. This
is illustrated in (13) where LF of (13a) is (13b) with the
reflexive reconstructed as a pronoun.
264
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
(13) John likes himself and Bill does too.
a) Johni likes himselfi, and Bill does too.
b) Johni likes[+a] himselfi, and Bill[-a] himi too.
Reconstruction refers to sentences with a reflexive inside a
moved NP (14) or sentences with a reflexive inside a moved
predicate (15).
(14) Johni wonders which pictures of himselfi/j Billj likes.
(15) How proud of herself*i/j does Maryi think that Nancyj is!
According to Chomsky (1993), reconstruction is relevant to
two functions at LF. The first option holds that only whelements undergo covert movement to wh-position at LF as
demonstrated in (16).
(16) a) John wounded [wh whichi] Bill saw [wh ti pictures of
himself].
b) John wondered [which x [Bill saw [x picture of
himself]].
Under this option, himself takes Bill as antecedent by
binding Principle A at LF.
The second option assumes that “which pictures of
himself” adjoins to “wh” (17a) so that the complementary
portions are deleted from the fronted phase and its copy.
This yields (17b) and its interpretation in (17c).
(17a) John wondered [wh which picture of himself] Bill saw
[wh which picture of himself]].
(b) John wondered [which picture of himself]i Bill saw [wh ti]].
(c) John wondered [which x, x a picture of himself [Bill saw x]].
265
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Under the second option, himself takes John as
antecedent by binding Principle A at LF. The available two
options at LF explain why himself can seek either John or Bill
as its antecedent.
Fiengo and May (1994) developed Kitagawa’s
suggestion and proposed a structural account under the
term “vehicle change”. They uphold that the strict reading
of reflexives involves a change from the reflexive to the
pronoun. This comes about as follows: a reflexive, when
copied from the first to the second clause, it is allowed to
change to pronoun. Thus, “vehicle change” allows strict
reading by reconstructing the reflexive as pronoun, which,
as set by Principle B of the Binding Theory, cannot be
locally bound. This explains why the reflexive in the overt
VP in (13), repeated here as (18a) can be reconstructed as
the pronoun in the elided VP in (18b), referring back to the
non-local subject NP John.
(18a) Johni likes himselfi, and Bill does too.
(18b) Johni likes[+a] himselfi, and Bill[-a] himi too.
Following the ongoing discussion, we realize that
reflexives in the VP-elliptical structures can be interpreted
sloppily as locally bound anaphora or strictly as pronoun.
This is in line with the Relevance Theory (henceforth RT) of
Sperber and Wilson (1995).
Relevance Theory
Relevance Theory (Sperber & Wilson 1995, 1998, 2002;
Wilson & sperber 2002) is an inferential theory of
communication, which aims to explain how the hearer
infers the communicator’s intended meaning. It holds that
human cognition is relevance-based (i.e. we pay attention to
information that appears relevant to us, construct relevant
representation of such information, and process these
266
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
representations in a context that maximizes its relevance).
Its principal claim is that the expectations of relevance
raised by an utterance are precise enough, and predictable
enough, to guide the hearer towards the communicator’s
meaning. In this respect, the Relevance Theory is broadly
construed as an inferential approach to pragmatics which is
based on a definition of relevance and two principles of
relevance - a Cognitive Principle (that human cognition is
geared to the maximisation of relevance), and a
Communicative
Principle
(that
utterances
create
expectations of optimal relevance). The goal of inferential
pragmatics is to explain how the hearer infers the speaker’s
meaning on the basis of the evidence provided.
According to Sperber (1994, 2000), all communicators
follow a path of least effort in computing cognitive effects,
but there are varying degrees of expectations of relevance in
the course of comprehension. He discusses three
sophisticated strategies used for comprehension. The
simplest strategy is what he calls naïve optimism (Sperber
1994:189). A hearer using this strategy looks for an
interpretation that appears to be relevant enough. If the
hearer finds one, he assumes that it was the intended
interpretation and attributes it to the communicator’s
meaning. A second strategy, a more complex one, is what
he calls cautious optimism (Sperber 1994:191). A hearer using
this strategy takes the first interpretation as relevant enough
and attributes it to the communicator’s meaning. Besides,
the hearer also considers what interpretation the
communicator “might have thought” would be relevant
enough (Wilson 2000). A third strategy is what he calls
sophisticated understanding (Sperber 1994:194). A hearer
using this strategy considers what interpretation the
communicator “might have thought he would think” was
relevant enough.
267
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Relevance, therefore, is treated as a property of inputs
to cognitive processes and analysed in terms of the notions
of contextual effect and processing effort. On the contextual
effect factor, the communicator guarantees that the
prepositions conveyed will be optimally relevant to the
hearer and that this relevant information will yield
adequate contextual effects to justify the hearer’s attention.
Therefore, contextual effect is achieved when newly
presented information interacts with the context of existing
assumption in one of the three ways: by strengthening an
existing assumption, by contradicting an existing
assumption or by combining with an existing assumption to
yield a contextual implication or a logical implication. It is
healthy to point out here that the contextual implication or
a logical implication is derived neither from the new
information alone, nor from the context alone, but from the
combination of the new information and the context. On the
processing effect factor, the communicator, who wants to
achieve adequate contextual effect, makes sure that the
utterance requires no more than the hearer’s minimum
justifiable processing effort. This is as a result of the fact that
“any increase in unjustifiable processing effort of the hearer
is an increase in the risk of misunderstanding” and “might
cause the overall relevance of the utterance to fall below the
acceptable level” (Wilson and Sperber 1991:588). Thus, to
Sperber and Wilson, communication has to do with the
hearer’s inference of the communicator’s intention.
In relevance-theoretic terms, utterances raise
expectations of relevance because the search for relevance is
a basic feature of human cognition. In this respect, an
utterance is relevant to an individual when it connects with
background information he has available to yield
conclusions that matter to him: say, by answering a
question he had in mind, improving his knowledge on a
certain topic, settling a doubt, confirming a suspicion, or
268
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
correcting a mistaken impression (Wilson and Sperber
2002). Hence, an input is relevant to an individual when its
processing in a context of available assumptions yields a
positive cognitive effect (Sperber & Wilson 1995). The
notion of a ‘positive cognitive effect’ is needed to
distinguish between information that merely seems to the
individual to be relevant and information that actually is
relevant. An efficient cognitive system is one which tends to
pick out genuinely relevant inputs; yielding genuinely true
conclusions (Sperber & Wilson, 1995).
The most important type of cognitive effect achieved
by processing an input in a context is a contextual
implication; a conclusion deducible from the input and the
context together, but from neither input nor context alone.
Relevance theory therefore claims that what makes an input
worth picking out from the mass of competing stimuli is not
just that it is relevant, but that it is more relevant than any
alternative input available to us at that time (Wilson &
Sperber 2002). The relevance of an input to an individual
depends on the positive cognitive effects achieved by
processing the input. For instance, the greater the positive
cognitive effects achieved by processing an input, the
greater the relevance of the input to the individual at that
time. On the other hand, the greater the processing effort
expended, the lower the relevance of the input to the
individual at that time. The following illustration, drawn
from Wilson & Sperber (2002), illustratively explains these
concepts. Mary, who dislikes most meat and is allergic to
chicken, rings her dinner party host to find out what is on
the menu. He could truly tell her any of three things:
(a) We are serving meat.
(b) We are serving chicken.
(c) Either we are serving chicken or (72 – 3) is not 46.
269
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
According to the characterisation of relevance, all three
utterances would be relevant to Mary, but (b) would be
more relevant than either (a) or (c). It would be more
relevant than (a) for reasons of cognitive effect: (b) entails
(a), and therefore yields all the conclusions derivable from
(a), and more besides. It would be more relevant than (c) for
reasons of processing effort: although (b) and (c) are
logically equivalent, and therefore yield exactly the same
cognitive effects, these effects are easier to derive from (b)
than from (c), which requires an additional effort of parsing
and inference (in order to work out that the second disjunct
is false and the first is therefore true). Thus, (b) would be
the most relevant utterance to Mary, for reasons of both
effort and effect.
As the foregoing discussion reveals, the Binding and
the Relevance theoretical frameworks are used for this
study to find out whether the interpretation of reflexives in
verb phrase elliptical sentences, by L2 learners of English in
Cameroon, is constrained by the minimal processing effort
or the syntactic order and the contextual effects.
Data and Methodology
The respondents for this study are 128 L2 learners of
English in Cameroon. The target population is made up of
university students, who are enrolled in the Department of
English of the University of Yaounde I, at different
proficiency levels (Level One (40), Level Two (45), and
Level Three (43)). These students come from different ethnic
background; thus, they speak different indigenous
languages and at least a lingua franca. They are chosen
because they have been exposed to English for at least
fourteen years (i.e. seven years in primary school and seven
years in secondary/high school). A production test was
deployed to test the subjects’ interpretation of reflexive
anaphora in VP-elliptical structures. The test had fourteen
270
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
test items and twenty-eight distracters. Six of the fourteen
experimental sentences were followed by a referential
context favouring strict interpretation as in (2) below. The
other eight sentences were in a null context. Each sentence
was followed by two alternative interpretations (i) and (ii)
(see samples below). The subjects were asked to check
which version matches their understanding of the
underlined part of the sentence. The following are samples
of the experimental sentences used in eliciting the subjects’
interpretation of reflexive anaphora in VP-elliptical
structures:
(1) Paul defended himself well and John did it too.
What does the underlined part of the sentence mean?
i) _______ John defended John.
ii) _______ John defended Paul
(2) Yvonne defended herself and Cynthia did too. Cynthia
was a good friend of Yvonne.
What does the underlined part of the sentence mean?
i) _______ Cynthia defended Cynthia
ii) _______ Cynthia defended Yvonne
(3) Mabel talked about herself and Deborah did too.
What does the underlined part of the sentence mean?
i) _______ Deborah talked about Deborah.
ii) _______ Deborah talked about Mabel.
(4) Max criticized himself and Jonathan did too. Jonathan did not like Max.
What does the underlined part of the sentence mean?
i) _______ Jonathan criticized Jonathan.
ii) _______ Jonathan criticized Max.
(5) Lucy praised herself and Jane did it too.
What does the underlined part of the sentence mean?
i) ______ Jane praised Jane.
ii) ______ Jane praised Lucy.
271
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Analysis and Results
This section probes into the responses provided by the
respondents in the data. It establishes the frequency of
occurrence of both strict and sloppy readings. This is done
to evaluate the extent to which the respondents adhere to
the binding principles in VP-elliptical structures.
Interpretation of reflexives in the referential context
favouring strict reading
Subjects were expected to have a strict reading of reflexive
anaphora in VP-ellipsis in a referential context favouring
strict reading. In considering sentences like [Yvonne
defended herself and Cynthia did to. Cynthia was a good
friend of Yvonne], the results in table 1 are recorded.
Table 1: Interpretation of reflexive anaphora in the referential context
favouring strict reading (n=6)
Level ONE
Level TWO
Level Three
Total
Score
%
Score
%
Score
%
Score
%
Sloppy reading
52
21.67%
37
13.70%
21
08.14%
110
14.32%
Strict reading
188
78.33%
233
86.30%
237
91.36%
658
85.68%
TOTAL
240
100%
270
100%
258
100%
768
100%
The table above shows that the subjects provided 110
(14.32%) responses in which they interpreted the reflexive
anaphora, in VP-elliptical sentences favouring strict
reading, sloppily; and 658 (85.68%) responses in which they
gave a strict reading as seen in example (19) below. This
gives a difference of 548 (71.35%) more strict reading.
19) Yvonne defended herself and Cynthia did too. Cynthia
was a good friend of Yvonne.
What does the underlined part of the sentence mean?
i) ____x___ Cynthia defended Cynthia. = (sloppy reading)
ii) ___x____ Cynthia defended Yvonne. = (strict reading)
272
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
The above statistics show that the respondents
interpret reflexive anaphora in VP-elliptical sentences, in
referential context favouring strict reading, both sloppily and
strictly. The results are presented on a mean percentage
graph below. It records the mean percentage of sloppy and
strict readings the respondents of each level of proficiency
provided in the interpretation of reflexive in VP-elliptical
structures.
Graph 1: Mean % graph of sloppy and strict readings in the referential
context favouring strict reading
100
80
60
Lev. 1
40
Lev. 2
20
Lev. 3
0
Sloppy
Strict
As can be inferred in the graph above, sloppy reading
diminishes as the subjects move up the academic ladder: Level
One respondents scored (21.67%), Level Two respondents
(13.70%) and Level Three respondents (08.14%). In the same
vein, strict reading progresses gradually as the subjects move
up the academic ladder: Level One (78.33%), Level Two
(86.30%) and Level Three (91.86%).
Interpretation of reflexives in the null context
It was expected here that subjects will go by Principle A of
the Binding Theory, which affects anaphors, to give a sloppy
reading of the reflexive anaphora in VP-elliptical structures.
In considering sentences like [Paul defended himself well
and John did it too]; [Mabel talked about herself and
Deborah did too], the results in table 2 were recorded.
273
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Table 2: Interpretation of reflexive anaphora in the null context (n=8)
Level One
Score
Level Two
Level Three
Total
%
Score
%
Score
%
Score
%
631
61.62%
Sloppy reading
170
53.12%
225
62.50%
236
68.60%
Strict reading
150
46.88%
135
37.50%
108
31.40%
393
38.38%
TOTAL
320
100%
360
100%
344
100%
1024
100%
As can be seen in the table, it turned out that the subjects
provided 631 (61.62%) responses in which they interpreted
the reflexive anaphora in VP-elliptical sentences sloppily;
and 393 (38.38%) responses in which they gave a strict
reading as seen in the example below.
20) Paul defended himself well and John did it too.
What does the underlined part of the sentence mean?
i) _____x__ John defended John.
= (sloppy reading)
ii) ____x___ John defended Paul.
= (strict reading)
The above statistics show that the subjects prefer more
a sloppy reading of the reflexive anaphora in VP-elliptical
sentences, in the null context, to a strict reading. The results
are presented on a mean percentage graph below. It records
the mean percentage of sloppy and strict readings the
subjects provided in the interpretation of reflexive in VPelliptical structures.
Graph 2: Mean % graph of sloppy and strict readings in the null context
80
60
Lev. 1
Lev. 2
Lev. 3
40
20
0
Sloppy
Strict
274
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
As can be inferred in the graph above, sloppy reading
progresses gradually as the subjects move up the academic
ladder: Level One (53.12%), Level Two (62.50%) and Level
Three (68.60%). In the same vein, strict reading diminishes
as the subjects move up the academic ladder: Level One
(46.88%), Level Two (37.50%) and Level Three (31.40%).
These results show that the minimal processing effort of the
Relevance Theory (RT) constrained these learners’
interpretation
of
reflexives
in
the
VP-elliptical
constructions. The respondents chose the most accessible
sloppy reading because it involves the least effort in
computing cognitive effects. Seeking the lower subject to be
the antecedent for the reflexive was easier than seeking the
higher subject because the distance between the reflexives
and the lower antecedent is shorter than the distance
between the reflexive and the higher subject. Furthermore,
the binding relationship between the reflexive and the
lower subject is within a single binding node as
demonstrated by the example below. The sentence “John
defended himself and Bill did too” has two binding nodes:
[IP John defended himself [conj and [IP Bill did too]]]
In contrast to the sloppy reading, the strict reading is
costly on the grounds that the pronoun reading, which
involves a change to the pronoun from the reflexive, is not
available in the linguistic information of the sentences.
Secondly, the distance between the reflexive and its higher
antecedent is greater than the distance between the reflexive
and its lower antecedent. Thirdly, the binding relationship
of the reflexive in the elided VP with its higher subject is not
within the single binding domain. Thus, constrained by the
least processing effort of RT, the L2 learners of English in
Cameroon, forged ahead with the most accessible sloppy
interpretation as evidenced by the high percentage of sloppy
reading.
275
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Furthermore, the results show that the interpretation of
reflexives in a null context, show that the minimal
processing effort of Relevance Theory constrained the
respondents’ interpretation of reflexives in the VP-elliptical
structures. This is evidenced by the high percentage sloppy
interpretation by all the proficiency levels: 53.12% for the
Level One, 62.50% for Level Two, and 68.60% for Level
Three. In all, the respondents scored 61.62% sloppy reading
as against 38.38% strict reading.
In the interpretation of reflexives in the referential
context favouring strict reading, it is realized that the
syntactic order and contextual effects of Relevance Theory
constrained the respondents’ interpretation of reflexives in
VP-elliptical sentences. This argument is supported by the
high percentage of strict reading of the reflexives in VPelliptical constructions favouring strict reading, in all the
proficiency levels: 78.33% for Level One, 86.30% for Level
Two, and 91.86% for Level Three. When the scores are
tallied, we notice that respondents scored 85.68% strict
reading as against 14.32% sloppy reading. Though the
contextual effects of Relevance Theory constrained the
respondents’ interpretation of reflexives in VP-elliptical
structures favouring strict reading, the respondents do not
reject sloppy reading out rightly in favour of strict reading
alone.
In all, both the sloppy and strict readings are adopted
by the respondents even in referential context favouring
strict reading. They interpreted the reflexives sloppily using
the strategy of naïve optimism which is constraint by the
least processing effort of Relevance Theory. In this regard,
we noticed that the respondents looked for the
interpretations that appeared to be relevant enough to them
at that time and assumed that they were the intended
interpretations.
As
such,
they
attributed
these
interpretations to the communicator’s meaning. They were
276
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
also adept in using the strategies of cautious optimism and
sophisticated understanding which resulted in strict
interpretation of reflexives. With regard to this, we could
say that the respondents looked for the interpretation that
appeared relevant to them in relation to context (contextual
effect), what the communicator might have thought would
be relevant to them, and what the communicator might
have thought they would think was relevant enough.
Finally, none of the respondents adopted a purely
sloppy or a purely strict reading of the reflexive anaphora in
VP-elliptical structure, although they were unanimous in
either a strict reading or a sloppy reading of a few structures.
The foregoing interpretation could be considered peculiar
in Cameroon English because of the influence of the
respondents’ first language (L1). Reflexive anaphora in the
various indigenous languages (first language (L1) of each of
the subjects) violate binding Principle A in both sides since a
reflexive anaphora in these languages can be bound outside
its local domain on the one hand and may be bound within
its local domain on the other hand. This is as a result of the
fact that the L1 of each of these respondents has both a
complex reflexive and a simplex reflexive. For instance,
there is a complex reflexive yi nyor (hisself) in Limbum (one
of the indigenous languages spoken in the North West
Region of Cameroon and by some of the respondents) and
echi-nyu (hisself) in Akɔɔse (one of the indigenous
languages spoken in the South West Region of Cameroon
and by some of the respondents) which binds long-distance
(i.e. to the subject of the higher clause) and a simplex
reflexive (e.g. nyor (self) in Limbum and nyu (self) in
Akɔɔse) which binds only locally. As a consequence of this,
it would not be erroneous to say that the phenomenon of
language transfer and/or influence may have played a
significant role here. However, findings reveal that
education affects the attainment levels in the interpretation
277
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
of reflexive anaphora in VP-elliptical constructions. There is
a significant correlation between the level of education and
the patterns of interpretation as far as the respondents are
concerned.
Conclusion
This study has investigated the interpretation of reflexive
anaphora in VP-elliptical sentences in Cameroon English.
The results indicate that both the principles of minimal
processing effort and the syntactic order and contextual
effects of the Relevance Theory, constrain L2 learners of
English in Cameroon in the interpretation of reflexive
anaphora in VP-elliptical sentences. Hence, they plow
ahead with both a sloppy and a strict reading of these
reflexives in the null context and in the referential context
favouring strict reading.
References
Chomsky, N. 1981. Lectures on Government and Binding. Dordrecht: Foris
publications.
Chomsky, N. 1993. “A Minimalist Program for Linguistics.” In The View
from Building 20: Essays in Linguistics in honor of Sylvain
Bromberger, ed Kenneth Hale and Samuel Jay Keyser, 1-52.
Cambridge, Mass.: MIT Press
Dalrymple, M., S. Shieber, and F. Pereira 1991. ‘Ellipsis and Higher-order
Unification’. Linguistics and Philosophy 14, 399-452.
Fiengo, R. and R. May 1994. Indices and Identity. Cambridge, MA: MIT
Press.
Finer, D., and E. Broselow 1986. “Second Language Acquisition of
Reflexive Binding”. In Proceeding of the sixteenth annual meeting of the
North Eastern linguistic society: 154-168.
Goldwater, M. and J. Runner 2006. “Coreference Interpretations of
Reflexives in Picture Noun Phrases: An Experimental Approach”. In
Proceedings of the 2004 Texas Linguistic Society Conference, (ed.) Pascal,
D. et al., 28-34. Somerville, MA: Cascadilla Proceedings Project.
Hestvik, A. 1995. ‘Reflexives and Ellipsis’. Natural Language Semantics 3,
211-237.
278
Napoleon Kang Epoge /Syllabus Review 3 (1), 2012 : 257 - 279
Kitagawa, Y. 1991. ‘Copying Identity’. Natural Language and Linguistic
Theory 9, 497-536.
Pan, H. 1998. “Closeness, Prominence and Binding Theory”. Natural
Language and Linguistic Theory 16: 771-815.
Reinhart, T. and E. Reuland 1993. “Reflexivity”. In Linguistic Inquiry 24:
657-720.
Sperber, D. 1994. Understanding Verbal Understanding. In J. Khalfa (Ed.)
What is intelligence? Cambridge: Cambridge University Press, pp. 17998.
Sperber, D. 2000. Metarepresentations in an Evolutionary Perspective. In
Sperber (ed.) (2000), 117-137.
Sperber, D. (ed.) 2000. Metarepresentations: An Interdisciplinary Perspective.
NY: OUP.
Sperber, D. and D. Wilson 1995. Relevance: Communication and Cognition.
Oxford: Blackwell.
Sperber, D. & D. Wilson 1995. Postface to the Second Edition of Relevance:
Communication and Cognition. Oxford: Blackwell.
Sperber, D. & D. Wilson 1998a. The Mapping between the Mental and the
Public Lexicon. In P. Carruthers & J. Boucher (eds.) Language and
Thought: Interdisciplinary Themes. Cambridge: Cambridge University
Press. 184-200.
Sperber, D. & D. Wilson 1998b. Irony and Relevance: A reply to Seto,
Hamamoto and Yamanashi. In Carston & Uchida (eds.), 283-93.
Amsterdam: John Benjamins.
Sperber, D. & D. Wilson 2002. Pragmatics, Modularity and Mind-reading.
Mind & Language 17. 3 -23.
Wilson, D. 2000. Metarepresentation in Linguistic Communication. In
Sperber (ed.) 2000, 411-448.
Wilson, D. & D. Sperber 2002. Truthfulness and Relevance. Mind 111. 583632.
279